An efficient starling murmuration-based secure web service model for smart city application using DBN

Abstract The advent of IoT devices increased internet usage more than smartphones and personal computers. The manual analysis of the Web Service Description Language (WSDL) document is quite expensive and time-consuming, hence this paper proposes a novel deep learning architecture to overcome the issue associated with web service classification. A structural self-organized deep belief network (SSODBN) is used for real-time web service classification in different fields such as Education, Smart electricity, Intelligent road networks, Health and social care, and Sports, water, and gas distribution. The SSODBN architecture utilizes a dropout strategy to minimize the interrelationship between the feature detectors and a regularized reinforced transfer function to eliminate the irrelevant weights. The main advantage offered by the S-DBN architecture is the improved preprocessing with feature selection. The Starling Murmuration Optimizer (SMO) is utilized in this paper to minimize the reconstruction error of the S-DBN architecture. The security of the smart city architecture is mainly improved via the blockchain defined network (BDN) using user-authenticated blocks. The experimental results revealed that the proposed method managed the scalability, latency, and centralization issues with superior data transmission.